1. Technical Field
This disclosure is concerned with improving the integrity and accuracy of detection apparatus, for example a cable locating instrument, which senses low frequency electromagnetic signals.
2. Discussion of Related Art
A cable locating instrument typically uses an array of sensors, arranged in a fixed geometry, to derive information about the relative direction of a buried utility. Low frequency magnetic sensors are a popular choice and work well when used in conjunction with a current source, which is connected to the utility.
A typical sensor uses a coil of wire wound on a ferrite core and is often referred to as an antenna. With such an antenna, it is possible to measure the magnetic field in one direction only or, by using a spatial array of antennas, a three-dimensional measurement is possible. Using two sensors on any single axis allows the magnetic field gradient to be measured. Using three sensors on any given axis allows the second-derivative of the magnetic field to be measured. Generally speaking using more sensors yields improved location accuracy, but this has to be set against the fact that increasing the number of sensors makes the equipment increasingly unwieldy.
It is important that the transfer functions of the antennas are well balanced between the channels. Without this balance the locator is disadvantaged for a variety of reasons:                Diminished sensitivity and accuracy of differential measurements.        Poor common mode rejection.        Poor depth accuracy.        Poor direction qualities from phase misalignment.        
There are various known ways of energizing the antennas and calculating a magnitude response at a number of predefined frequencies. In one known system a small dipole generator known as a micro-sonde is positioned on the magnetic axis between two ferrite-cored sensors to emit an alternating magnetic field. Ideally the sonde would be positioned at the exact mid-point between the antennas and perpendicular to the magnetic axis. The sonde can be energized at a number of frequencies allowing the magnitude response from the two sensors to be compared.
This system provides an adequate self-check mechanism but nothing more. Various errors can easily be introduced:                Very small changes in relative position of the micro-sonde to the sensors will cause a significant change to the magnetic flux linkage.        If the locator is near a ferrous magnetic material (e.g. cast iron) then the entire magnetic aperture of both sensors will be distorted.        It is difficult to energize the micro-sonde without causing magnetic field spillage from the energizing windings. These may produce unwanted and ill-defined components to the resulting magnetic field.        
In another known system the ferrite sensors each have an additional self-test winding inductively coupled to the main sensor winding and which is energized via a frequency programmable current source. This is an improvement to the micro-sonde system since the programmable current source can separately energize both antenna sensors through a multiplexer switch. The coupling to the antenna is a parallel winding which gives a strong flux-linkage when wound over the central part of the ferrite. This system yields two signals which can be measured: (i) the emf induced in the antenna which is energized, and (ii) a smaller emf induced in the other antenna caused by the radiated magnetic field and electro-magnetic induction in the opposite sensor. This self-test allows four measurements to be made at each frequency under test, which yields an improvement. However, drawbacks of this system are:                The additional self-test coil wound on each sensor produces additional parasitic capacitance which itself modifies the transfer function of the sensor—typically the self-resonant frequency is lowered and this is generally unhelpful.        Like the micro-sonde system, it is highly susceptible to micro movement between the sensors, e.g. expansion due to temperature.        Like the micro-sonde system it is highly susceptible to changes in the magnetic permeability close to the antennas.        
In summary therefore, there are imperfections in the self-test measurements possible for both systems and as a result only the magnitude responses are calculated and compared. The standard deviation of a given sample of measurements is too high to be regarded as a high integrity check. Accordingly these systems can only be regarded as basic checks—not instrumentation quality.
To understand the accuracy requirements it is necessary to consider the geometry of a magnetic field. As an example, a 2% error in the calibration balance of top and bottom antennas in a cable locator with 35 cm separation between the antennas can be shown to result in a depth calculation error of at least 37.5%.
There are many other undesirable consequences of a poor calibration balance. The differential sensitivity (bottom-top) is compromised as is the ability to reject a signal which is common to both bottom and top since by definition there will be an unbalance proportional to the calibration error.
Therefore, there is a need to improve detection accuracy, increase detection speed and reduce hardware requirements.